GENBAND US LLC, Plaintiff-Appellant v. METASWITCH NETWORKS CORP., METASWITCH NETWORKS LTD., Defendants-Appellees
Appellant Genband US LLC sued Metaswitch Networks Corp. and Metaswitch Networks Ltd. (together, Metaswitch) for patent infringement. After a jury found that Metaswitch infringed various claims of several of Genband's patents, and that the claims at issue had not been proven invalid, Genband sought a permanent injunction. The district court denied the request, concluding that Genband had not established irreparable harm from the infringing activities. That conclusion, however, may have relied on too stringent an interpretation of the requirement, for an injunction, that the allegedly irreparable harm is being caused by the infringement. Based on the district court's opinion and the briefing in this court, moreover, we cannot be confident of the answer to the causation question under the standard properly governing the inquiry or whether there is any independent ground for finding no irreparable harm or otherwise denying an injunction. Accordingly, we vacate the denial of the injunction and remand for reconsideration.
Genband sells products and services that help tele-communications companies offer voice-communications services over Internet Protocol networks, i.e., “voice over IP” (VoIP) services. Genband owns a number of patents related to its offerings, some of them acquired in 2010 when it purchased the Carrier VoIP and Application Solutions line of business from Nortel Networks Inc. out of Nortel's bankruptcy. Metaswitch sells telecommunica-tions products and services that compete with Genband's offerings, though Metaswitch was not a major competitor until recent years. See Genband US LLC v. Metaswitch Networks Ltd, 211 F. Supp. 3d 858, 865–66, 871–72 (E.D. Tex. 2016).
On January 21, 2014, Genband filed a complaint against Metaswitch in the United States District Court for the Eastern District of Texas, alleging that certain Metaswitch products infringed and/or continue to infringe seven U.S. patents owned by Genband: U.S. Patents Nos. 6,772,210; 6,791,971; 6,885,658; 6,934,279; 7,995,589; 7,047,561; 7,184,427; and 7,990,984. See id. at 866–69. The lawsuit proceeded to a jury trial in January 2016, and the jury found that Metaswitch infringed all asserted claims and that those claims were not invalid. Id. at 868. The jury awarded $8,168,400 in damages. Id.
The district court thereafter held a bench trial to ad-dress various matters, including equitable defenses and Genband's request for a permanent injunction. Id. On September 29, 2016, the district court issued an opinion and order containing extensive findings of fact and ac-companying conclusions of law. Among other things, the court rejected Metaswitch's equitable defenses, including laches, finding no unreasonable delay by Genband in asserting the patents. Id. at 895–901.
The court also denied Genband's request for a permanent injunction. Id. at 894–95. The district court rested its denial entirely on the determination that Genband failed to show that it would suffer irreparable harm from Metaswitch's continued infringement. The court gave two reasons, without indicating that the second reason inde-pendently supported its determination.
First, the court held that Genband did not demonstrate a causal nexus between the alleged irreparable harm (based on lost sales) and the presence of the infringing features in Metaswitch's infringing products. Id. at 894–95. In so ruling, the district court stated that “it is Genband's burden to demonstrate that the patented features drive demand for the product.” Id. at 894. The court borrowed certain language from this court's decision in Apple, Inc. v. Samsung Electronics Co. (Apple II), 695 F.3d 1370, 1375 (Fed. Cir. 2012) (“The patentee must ․ show that the infringing feature drives consumer demand for the accused product.”), which in turn relied on similar language in Apple, Inc. v. Samsung Electronics Co. (Apple I), 678 F.3d 1314, 1324 (Fed. Cir. 2012) (“If the patented feature does not drive the demand for the product, sales would be lost even if the offending feature were absent from the accused product.”). Before reiterating the “drive demand” principle, the district court quoted this court's statement in Apple, Inc. v. Samsung Electronics Co. (Apple III), 735 F.3d 1352, 1364 (Fed. Cir. 2013), that “this inquiry should focus on the importance of the claimed invention in the context of the accused product, and not just the importance, in general, of features of the same type as the claimed invention.” Genband, 211 F. Supp. 3d at 894. The district court then noted Genband's arguments that certain stringency-reducing explanations of “drive demand” are found in both Apple III, 735 F.3d at 1365 and Apple, Inc. v. Samsung Electronics Co. (Apple IV), 809 F.3d 633, 641–42 (Fed. Cir. 2015), cert. denied, 136 S. Ct. 2522 (2016), petition for cert. filed, 85 U.S.L.W. 3460 (U.S. Mar. 10, 2017) (No. 16-1102). But the court did not indicate agreement with Genband that those explanations state the governing law. Genband, 211 F. Supp. 3d at 894.
The court then applied its articulated legal standard as follows:
During the bench trial, Genband presented the following regarding the causal nexus, which falls into three general categories: (1) a self-generated “win-loss” report; (2) demonstratives purporting to correlate dates of Metaswitch press releases with an alleged decline in Genband's market share; and (3) statements from Metaswitch marketing materials and opinion tes-timony from Mr. McCready [a Genband execu-tive].
Genband's presentation of evidence does not satisfy its burden to show causal nexus. Accordingly, Genband fails to show that it has suffered irreparable harm as required for a permanent injunction.
Id. at 894–95.
The district court's second reason for finding no irreparable harm involved Genband's litigation choices. The court found that, although Genband did not unreasonably delay in suing Metaswitch for infringement, it did delay in suing for several years after analyzing Metaswitch's products, and the court also observed that Genband did not seek a preliminary injunction. Those facts, the court concluded, weighed against a finding of irreparable harm from Metaswitch's sales. Id. at 895. The district court denied the requested permanent injunction without addressing other considerations. Id.
Genband appeals. We have jurisdiction under 28 U.S.C. § 1292(a), (c)(1).
We review a district court's grant or denial of injunc-tive relief for abuse of discretion. Ecolab, Inc. v. FMC Corp., 569 F.3d 1335, 1352 (Fed. Cir. 2009). “A district court would necessarily abuse its discretion if it based its ruling on an erroneous view of the law or on a clearly erroneous assessment of the evidence.” Cooter & Gell v. Hartmarx Corp., 496 U.S. 384, 405 (1990); see Highmark Inc. v. Allcare Health Mgmt. Sys., Inc., 134 S. Ct. 1744, 1748 n.2 (2014). A “clear error of judgment in weighing relevant factors” is also a ground for finding an abuse of discretion. Ecolab, 569 F.3d at 1352 (quoting Innogenet-ics, N.V. v. Abbott Labs., 512 F.3d 1363, 1379 (Fed. Cir. 2008)). Where it is not evident that a district court has applied the correct legal standard in exercising its discretion, we may vacate and remand for the district court to do so in the first instance, especially where further factual findings may be warranted under the correct legal standard. Id.; see Apple III, 735 F.3d at 1368.
In this case, the sole basis for denial of the requested injunction was the district court's finding that Genband did not show irreparable injury from the conduct it sought to enjoin, one precondition to issuing the requested injunction. See eBay Inc. v. MercExchange, L.C.C., 547 U.S. 388, 391 (2006). Genband relied on evidence that Metaswitch was making sales in direct competition with it, causing Genband to lose sales and thereby to suffer harms of the type often found irreparable. See Apple IV, 809 F.3d at 640–41; Broadcom Corp. v. Emulex Corp., 732 F.3d 1325, 1336–37 (Fed. Cir. 2013); Douglas Dynamics, LLC v. Buyers Prods. Co., 717 F.3d 1336, 1345 (Fed. Cir. 2013). But the district court held that Genband had not met a requirement that is part of the irreparable-injury component of eBay in cases like this—namely, the requirement of “some causal nexus” between the infringing features of the infringer's products and the sales lost to the patentee. Apple I, 678 F.3d at 1324; see Apple II, 695 F.3d at 1374–75 (“a sufficiently strong causal nexus [that] relates the alleged harm to the alleged infringement” is “part of the irreparable harm calculus”); Apple III, 735 F.3d at 1364 (“some causal nexus between [defendant's] infringing conduct and [patentee's] alleged harm” is required); Apple IV, 809 F.3d at 640 (requiring “a causal nexus linking the harm and the infringing acts” to ensure that “an injunction is not entered on account of ‘irreparable harm caused by otherwise lawful competition’ ” (quoting Apple III, 735 F.3d at 1361)).
The district court's opinion, however, leaves us uncer-tain whether the court relied on too stringent an interpretation of the causal-nexus requirement. The court declared that Genband had to prove that “the patented features drive demand for the product.” Genband, 211 F. Supp. 3d at 894; see id. (quoting Apple II's reference to “drives consumer demand”). But we cannot be sure that the district court, in demanding such proof, used the standard for causal nexus now established to be the governing standard.
The “drive demand” formulation, on its face, is sus-ceptible to importantly different interpretations, some stricter, some more flexible, at least in situations where the product at issue has multiple purchasers and multiple features that different purchasers might assign different weights in their purchasing decisions. For example, as the district court in Apple III had assumed, the “drive demand” formulation could require that the infringing feature be “the driver” of decisions by consumers treated collectively as a kind of unit, even requiring proof that no or almost no buyers would buy the product but for the infringing feature. Or it could require less, e.g., that the infringing feature be “a driver” of decisions by a substan-tial number of individual consumer decision-makers considering multiple features.
Here, Genband argued for a standard on the less stringent side of the spectrum. The district court de-scribed Genband's argument, but the court did not itself say anything to indicate its adoption of the argument. Genband, 211 F. Supp. 3d at 894. Yet it has been clear since at least Apple III that a standard of the less demanding variety—as an interpretation of “drive demand,” a standard based on “a driver” as opposed to “the driver,” applied in the multi-consumer, multi-feature context—is the governing one for what suffices to meet the causation component of the requirement of irreparable injury, i.e., that the injury asserted to be irreparable be injury from the defendant's use of infringing features.1
Even in Apple I, which used the phrase “drive the demand,” the court articulated the governing standard as “some causal nexus” or “a nexus.” 678 F.3d at 1324, 1327. And what the court declared insufficient to establish a causal nexus was a “mere showing that Apple might lose some insubstantial market share as a result” of the infringement, i.e., that “the alleged infringement caused an insignificant amount of lost sales.” Id. at 1324 (emphases added). In Apple II, the court found insufficient evidence to establish the required causal nexus where “the only pertinent evidence” showed that the infringing feature was “not one of the top five reasons consumers select the” allegedly infringing product. 695 F.3d at 1376 (“[T]he causal link between the alleged infringement and consumer demand for the [accused device] is too tenuous to support a finding of irreparable harm.”). The court explained that “[i]t is not enough for the patentee to establish some insubstantial connection between the alleged harm and the infringement and check the causal nexus requirement off the list.” Id. at 1375.
In Apple III, the court repeated the “some causal nexus” standard and rejected certain formulations as too stringent. 735 F.3d at 1364. It stated that the patentee need not “show that one of the patented features is the sole reason consumers purchased” the accused products. Id.; see id. (ruling that patentee need not “show that a patented feature is the one and only reason for consumer demand” and explaining that such a requirement is too rigid and inflexible given the complexity of consumer preferences). And it stated that “a showing of causal nexus does not require” proof “that consumers will buy [the accused product] instead of [the patentee's competing product] because it contains [the infringing] feature.” Id. at 1367 (internal quotation marks omitted); see id. at 1364 (same). More affirmatively, the court explained:
Thus, rather than show that a patented feature is the exclusive reason for consumer demand, [the patentee] must show some connection between the patented feature and demand for [the infringer's accused] products. There might be a variety of ways to make this required showing, for example, with evidence that a patented feature is one of several features that cause consumers to make their purchasing decisions. It might also be shown with evidence that the inclusion of a patented feature makes a product significantly more desirable. Conversely, it might be shown with evidence that the absence of a patented feature would make a product significantly less desirable.
Id. at 1364; see also id. at 1367 (referring to whether “the patented features influenced demand for [the infringer's] products”); id. at 1368 (stating that “evidence that a feature significantly increases the desirability of a product incorporating that feature” may show that the feature “drives demand”).
In Apple IV, the court stated that the causal-nexus requirement “just means that there must be proof that the infringement causes the harm” that the patentee alleges is irreparable—in that case, “damage to [the patentee's] reputation as an innovator, lost market share, and lost downstream sales.” 809 F.3d at 639. The court reiterated the error of a “sole cause” standard, especially where multiple features are present in the product at issue, id. at 641, explaining that the district court in that case had erred in requiring the patentee “to prove that the infringing features were the exclusive or predominant reason why consumers bought [the accused] products to find irreparable harm,” id. at 642 (emphasis added).
Instead, the court stated, borrowing from Apple III, “proving a causal nexus requires the patentee to show ‘some connection’ between the patented features and the demand for the infringing products.” Id. at 641 (quoting Apple III, 735 F.3d at 1364). The court explained this requirement as demanding “evidence that ‘a patented feature is one of several features that cause consumers to make their purchasing decisions.’ ” Id. at 642 (quoting Apple III, 735 F.3d at 1364). In the case before it, the court explained, “[t]he district court should have deter-mined whether the record established that a smartphone feature impacts customers' purchasing decisions.” Id. at 641; see id. at 641 n.1 (discussing various examples). And in ruling that irreparable harm was proved in the case before it, the court relied on its conclusion that the record established that the patented features “do influence consumers' perceptions of and desire for these products.” Id. at 642.
In the present case, Genband specifically invoked the standards laid out in Apple III and Apple IV that we have just summarized. The district court, however, referred to those standards only by stating that Genband “argue[d]” for them. Genband, 211 F. Supp. 3d at 894. In these circumstances, we see no sufficient basis for inferring that the district court actually used those standards, rather than an unduly stringent test, to interpret and apply the “drive demand” standard.
The clarified standards set forth in Apple III and Apple IV govern the causal-nexus inquiry, at least in a multi-purchaser, multi-component situation in which only a component of a larger product or system is covered by the patent in suit.2 The formulations in those decisions avoid a too-demanding causal-nexus requirement that might be attributed to the “drive demand” language. The standard prescribed by Apple III and Apple IV, as appro-priate to the multi-purchaser, multi-component context, lies between the unduly stringent “sole reason” standard we rejected in Apple III and Apple IV and the unduly lax “insubstantial connection” standard we rejected in Apple II. The standards seek to reflect “general tort principles of causation,” Apple III, 735 F.3d at 1361, and to make proof of causal nexus practical “from an evidentiary standpoint,” Apple IV, 809 F.3d at 641, in a multi-purchaser, multi-component setting.
Where the patentee relies on lost sales to show irreparable injury, it matters what reasons various buyers have for making the purchases lost to the patentee. If all but an insignificant number of purchases from the infringer would have been made even without the infringing feature, the causal connection to the asserted lost-sale-based injury is missing. But this court's cases have now made clear that, under the causation approach suitable for a multi-feature, multi-purchaser context, the patentee may be able to make the causal connection between infringement and the relevant lost sales through evidence of various kinds, e.g., that the infringing features significantly increased the product's desirability, that soundly supports an inference of causation of a significant number of purchasers' decisions.3
Of course, the causation requirement does not end the injunction inquiry, even as to the irreparable-injury requirement, let alone as to the other elements of the eBay analysis. But here the only dispositive basis of the district court's denial of the injunction was the causal-nexus requirement. And we cannot be confident that the district court applied the current governing approach to causation rather than an unduly demanding approach.
We conclude that a remand is needed. We are not in a position to conclude that applying the Apple III/Apple IV standards would make no difference to the district court's finding of no causal nexus and, hence, no irreparable injury. In its application of the “drive demand” formulation, the district court included just one paragraph, making only a summary reference to Genband's evidence, without explaining in that paragraph why that evidence was deficient. See id. at 894–95. And in this court, Genband has not only argued about the evidence the district court mentioned in that paragraph but also point-ed to extensive additional evidence, not discussed in that paragraph, as relevant to the inquiry. Given the roles of fact-finding and discretion in the inquiry, it is for the district court, not for this court, to undertake application of the proper causal-nexus standard to the full record in this case.
Apart from its causal-nexus determination, the district court deemed the timing of Genband's suit and Genband's choice not to seek a preliminary injunction to weigh against a finding of irreparable injury. Genband asks us to disapprove of the court's analysis of those considerations. Genband correctly points out that, when a patent owner postpones suit and forgoes a preliminary injunction, there may well be reasons for the patent owner's actions independent of any implied concession that the infringement-caused injury is not actually irreparable: for example, the competitive threat may initially be small, or the merits may be much better presented through full litigation than through abbreviated preliminary-injunction proceedings. See, e.g., Petrella v. Metro-Goldwyn-Mayer, Inc., 134 S. Ct. 1962, 1976 (2014); Mytee Prods., Inc. v. Harris Research, Inc., 439 F. App'x 882, 888 (Fed. Cir. 2011). But Genband has not justified a per se rule making the patent owner's choices about when to sue and whether to seek interim relief legally irrelevant.
In this case, the timing of Genband's suit and Genband's decision not to ask for preliminary relief call for an evidentiary judgment—a determination of what weight they have in determining irreparability of the harm at issue (under the governing legal standards) in the context of the evidence as a whole. We are remanding for a redetermination of the causal-nexus issue. That determination, and the findings made in making it, may affect the need for and content of the required evidentiary evaluation of these additional, irreparability considerations. We therefore include these matters in the remand.
For the foregoing reasons, we vacate the denial of the motion for a permanent injunction and remand for reconsideration consistent with this opinion.
Costs awarded to Genband.
VACATED AND REMANDED
1. We are concerned here only with standards in the injunction context. “Drive demand” and related formulations appear in damages cases, but that context involves distinctive apportionment-based, evidentiary-reliability considerations. See, e.g., Commonwealth Sci. & Indus. Research Org. v. Cisco Sys., Inc., 809 F.3d 1295, 1301–04 (Fed. Cir. 2015); LaserDynamics, Inc. v. Quanta Comput., Inc., 694 F.3d 51, 67 (Fed. Cir. 2012); Rite–Hite Corp. v. Kelley Co., 56 F.3d 1538, 1549 (Fed. Cir. 1995) (en banc).
2. The causal-nexus inquiry may have little work to do in an injunction analysis when the infringing product contains no feature relevant to consumers' purchasing decisions other than what the patent claims. In such a case, causal nexus and consumer demand may be appar-ent from the simple fact of infringing sales.
3. We have not had, and do not here have, occasion to undertake further refinements, e.g., by developing rules for proof sequencing of the sort found in other areas of law, such as in, to cite just the most recent example, Maslenjak v. United States, No. 16-309, 2017 WL 2674154, at *8–10 (U.S. June 22, 2017) (describing proof scheme for causal influence of falsehood in obtaining citizenship).